There is a moment familiar to anyone in a family where frontotemporal dementia runs in the bloodline: a form on a kitchen table asking about your health, and about your parents'. Not a scan, not a clinic. A form. When Charlotte Graafland and colleagues at Erasmus University Medical Center in Rotterdam sat down with 28 Dutch adults from families carrying pathogenic variants in GRN, MAPT or C9orf72 and asked them why they had, or had not, taken a presymptomatic genetic test, the answers included the ones you would expect, and then some you might not. People spoke about the psychological weight of a positive result, about holding on to hope, about not wanting to know something they could do nothing with. They also spoke about mortgages, about life and disability cover, and about what an employer might make of it. That study, published online in Human Genetics on 16 February 2026, is a qualitative one: a secondary analysis of semi-structured interviews, examined with inductive thematic analysis. It reports what people said, not how many people in the population say it.
The sample is worth stating plainly, because it sets the boundary of what can be claimed. Thirty people were invited and two declined; 28 took part, 12 recruited through the Department of Clinical Genetics and 16 through the FTD-RisC cohort. Eleven already knew they carried a variant, three knew they did not, and 14 were at 50 percent risk and untested. Reasons in favour of testing clustered around wanting to know and relieving uncertainty, hoping to test negative, planning a life, making reproductive decisions, explaining future symptoms and contributing to the common good. Reasons against clustered around fear of the psychological impact, fear of effects on employment, insurability or a mortgage, maintaining hope, and a sense that a result had no immediate use, partly because nobody can tell a carrier when symptoms might begin. The authors note that people who tested and people who did not weighed the same considerations differently, and that cohort studies such as FTD-RisC and the European GENFI consortium, which do not require participants to learn their genetic status, made knowing feel less urgent to some: the regular scans and blood draws functioned as reassurance.
Here is where the story stops being about counselling rooms. In the Netherlands, where these interviews were done, the Medical Examinations Act sets a question threshold: below a total insured sum of 352,200 euros in life cover over three years, and below 51,100 euros for the first year of disability cover and 34,200 euros thereafter, insurers may not ask about hereditary testing in you or your relatives. Those figures took effect on 1 January 2025. Above the threshold, the questions are permitted. Dutch mortgages are not small, and a term life policy attached to one can cross that line. In the United Kingdom the arrangement is different again: under the Code on Genetic Testing and Insurance, a voluntary agreement between the government and the Association of British Insurers first published in October 2018, only one predictive test can be asked about at all, for Huntington's disease, and only for life cover above 500,000 pounds. Test results obtained through research participation do not have to be disclosed. The government's second three-year review, published on 5 March 2026, records that no insurer has applied to add a condition to that list in more than twenty years. In the United States the gap is wider: the Genetic Information Nondiscrimination Act reaches health insurance and employment, but not life, disability or long-term care cover, a limit that a 2021 survey of the general public by Prince and colleagues in Risk Management and Insurance Review found is poorly understood by the people it does and does not protect.
The reason this matters beyond the individual family is arithmetic. Every therapy in development for genetic FTD needs people who know their genotype. A scoping review of three trial registries published in NeuroSci in November 2025 identified twelve registered interventional trials aiming at disease modification in FTD, half of them targeting GRN. The largest read-out so far, INFRONT-3, was designed to enrol GRN carriers who were symptomatic and carriers who were at risk but not yet symptomatic; Alector announced on 21 October 2025 that the 96-week trial did not meet its clinical co-primary endpoint, results reported by the company and not yet peer reviewed. Prevention, if it ever arrives, will require enrolling people years before anything is wrong with them, which means enrolling people who have chosen to find out. A separate UK qualitative study of fourteen people connected to familial FTD, published in Molecular Genetics and Genomic Medicine in November 2024, found the same tangle of personal, practical and systemic considerations around trial participation.
So the deterrent that shows up in these interviews is not written into anyone's DNA. It is written into statute, into voluntary codes, into the small print of a health declaration, and it differs depending on which country you happen to raise your children in. A family in Rotterdam, a family in Rochester and a family in Reading face the same 50 percent risk and three quite different sets of consequences for knowing. None of this tells any individual whether to test; that conversation belongs in a clinical genetics service, with time and support around it. What the Dutch interviews do is name the part of the decision that a health system cannot counsel away, because it is not a fear. It is a rule.